Recent advances in high-ISO digital technology have greatly expanded both practical and artistic photographic capabilities, allowing casual photographers to readily capture images and data previously unobtainable with a hand-held camera.
Practical uses include photo-graphing construction details in the dark recesses of a troubled project, distant wildlife at first light and industrial operations at night. Artistically, good low-light capabilities provide new ways to visualize and depict the world around us rather than merely imitating the past.
After some trial and much error, I’ve evolved a personal low-light exposure and post-processing technique that relies on the high dynamic range of RAW image files from specific cameras to produce large, high-quality, fine-art prints. Fine-art prints are usually at the apex of output quality, so this same technique should be workable for day-to-day practical applications, as well. Your mileage may vary, of course.
The low-light exposure concepts that I’ll discuss today, and the post-processing software techniques that I’ll discuss next week, work well for me. In the Web version of this article, which you can find by pointing your Internet Web browser to http://www.redoubtreporter.wordpress.com, we’ve posted several examples from my new solo photo exhibit, “Veiled Images,” which opens at Kenai Peninsula College’s Kachemak Bay Campus gallery Feb. 27. All images in the exhibit were shot hand-held at ISO 1,600 to ISO 3,200 in very dim light, and then printed quite large, about 23 inches on the long side. Next week, we’ll take a look at the specific post-processing techniques used to optimize these photos.
Overexposure and flat contrast are among the most common problems encountered in low-light photography, aside from camera shake and subject motion blurring. Almost by definition, low-light situations really do tend to be dark, with deep shadows, a few too-bright highlights and poor separation of the midtones. Photos of these situations should not be unnaturally bright if they’re to look true to life, yet we need to preserve contrast between adjacent midtones and not lose detail in the brighter areas. Exposure tailored to facilitate later post-processing is crucial.
All cameras are designed to meter and expose scenes as if they were a uniform 18 percent middle gray, so overexposure often occurs when large, dark areas are metered and exposed to be brighter than they really are. When overexposing an image, a too-long shutter speed is a frequent culprit, and that too-long shutter speed also causes objectionable blurring.
Using shutter-speed priority mode (usually labeled as S or T mode), rather than the more common Aperture-priority (A) mode, helps control blurring, allowing you to specify a fast-enough shutter speed. For static subjects like the sample photos posted with this week’s article, set the shutter speed no slower than the slowest speed that you know from previous tests you can hand-hold steadily without any blur. Set a balanced ISO and let the aperture fall where it will. Potential optical softness at wider lens apertures will be the least of your problems.
When making pictures of people in social settings, assume that a shutter speed of 1/8 second or shorter will be required to avoid subject motion blurring of relatively static people who are sitting and conversing. A shutter speed of at least 1/30 second or even shorter will be needed to minimize motion blurring of moving subjects, such as people gesturing. A fast-enough shutter speed may require high-ISO settings even when the light seems fairly bright. You can tame the resulting image noise with later post-processing.
A lens that’s both bright and sharp at wide apertures helps a great deal under these circumstances. My favorite lenses in low-light situations are the moderately wide-angle Panasonic 20-mm f/1.7 and Olympus’ 12- to 40-mm f/2.8 PRO zoom lens, used on a five-axis-stabilized Olympus OM-D camera body. Both lenses, particularly the Panasonic, are sharp and contrasty, even at very wide apertures.
Sigma’s 35-mm f/1.4 Art series prime lens and 18- to 35-mm f/1.8 Art series zoom lens should work comparably well for a variety of full-frame and APS-C cameras, although they’re larger, heavier and cost more. Shorter focal lengths have decent depth of field at wide apertures, so they’re particularly useful when taking photos of indoor social gatherings and in other tight quarters.
I compensate for both overexposure and blur by relying on the camera’s inherent dynamic range to use faster shutter speeds and “underexpose” the photos. Later, I’ll adjust those apparently underexposed images with post-processing software to brighten apparently too-dark areas. It’s generally easier to correct underexposed images and recover detail from them than from overexposed images with featureless highlights.
I’ll set the camera’s exposure compensation to reduce the metered exposure by -.7 EV, while also setting the camera to make three bracketed exposures each time I press the shutter release, changing each bracketed exposure by +/- .7 EV. That bracketed setting results in one exposure that’s .7 EV darker than the metered 18 percent middle gray, one even darker exposure that’s -1.3 EV darker than metered, and one brighter exposure that matches the camera’s 18 percent neutral gray metering.
Because dim-light situations are so variable and difficult to “eyeball,” bracketing greatly increases your probability of getting at least one exposure that’s close enough to make a good print after careful post-processing. With adequate dynamic range, a decent print can often be made from files that are two EV “underexposed” relative to the camera’s metered exposure. Don’t forget to make several bracketed shots of each image to ensure that at least one set is not blurred.
This technique works reasonably well with RAW image files made with cameras using modern high-dynamic-range Sony sensors. I’ve found that files made with Olympus OM-D, E-P5 and E-PL5 Pen cameras, Pentax K-5 models and Nikon D5200, D5300, D5500 and D7000 cameras all work well with this technique. These affordable Nikon and Pentax APS-C sensor cameras have excellent dynamic range, as much as 14 EV, while smaller Micro Four-Thirds Sony sensors have a dynamic range of at least 12.3 EV, not quite as good, but good enough with proper post-processing. I don’t want to offend Canon owners, but tests that I’ve seen as recently as Feb. 16 consistently indicate that current Canon APS-C cameras remain impeded by lower dynamic range.
Using RAW format image files is critical to high-quality low-light photography. RAW image files made with these and comparable cameras have a great deal of recoverable detail hidden in apparently dark areas while maintaining good highlight detail. JPEG files have a very limited dynamic range, only 8 EV, and “bake” the final file in a way that’s virtually impossible to correct later in post-processing.
We’ll conclude next week with a discussion of Adobe Lightroom and DXO post-processing techniques to make both specific and overall corrections that optimize low-light RAW images.
Local attorney Joe Kashi received degrees from MIT and his law degree from Georgetown University. He has published articles about computer technology, law practice and digital photography in national media since 1990. Many of his articles can be accessed through his website, http://www.kashilaw.com.